Data Model Wizard
- 5 minutes to read
XPO provides the Data Model Wizard that creates a data model mapped to an existing or newly created database. The model created includes XPO classes and metadata, and can be modified in the Data Model Designer. The wizard also creates the static ConnectionHelper class that exposes an auto-created connection string and provides methods that simplify the Data Access Layer and data store initialization. This topic demonstrates how to launch the wizard and describes wizard pages.
To invoke the wizard, use the DevExpress ORM Data Model Wizard‘s built-in project item template.
Choose “Database First” or “Model First”
The following page is displayed when the Data Model Wizard is invoked.
On this page, you can either choose to use the existing database (the Database First workflow) or create a new database (the Model First workflow).
| Option | Description |
|---|---|
| Map to an existing database | The wizard asks you to specify a connection to an existing database. The database structure will be reverse engineered and the data model will be generated automatically. |
| Create a new database | The wizard will ask you to specify the connection to the database that will be used in your application. You can point to a non-existent database - XPO can automatically create it later (see Tutorial 1 - Your First Data-Aware Application with XPO). |
| Do not connect to a database | The wizard will not configure a database connection. This option can be useful if you develop a persistent class library that is not bound to any database and can be used in different applications. In XAF applications, choose this option, because XAF manages the database connection itself. |
Select the desired option and click Next.
Specify the Database Connection
On the second page, you can specify database connection settings.
Choose the provider, and then specify provider-specific options. The image above illustrates how to set up a connection to a database hosted by a local instance of the Microsoft SQL Server and accessed using Windows Authentication.
The Import stored procedures option is available when you follow the Database First workflow. When checked, the wizard will generate auxiliary helper classes that allow you to directly call existing stored procedures and handle the results.
Note
If the wizard throws an error such as “Could not load type MySql.Data.MySqlClient.MySqlConnection“, “Could not load type NpgSql.NpgSqlConnection“, unable to load DLL ‘fbembed’ etc., then ensure that you have the required database provider assembly installed. Assemblies required by XPO data store adapters are listed in the Database Systems Supported by XPO topic. You need to put the required DLLs in a place where they can be found by the Visual Studio process. You can install .NET assemblies to the Global Assembly Cache (GAC) and copy the required native DLLs (if any) to %SystemRoot%\System32, to make them available globally.
Database First
If you chose the Create a new database option on the first page, then the wizard will complete after specifying the database connection, and the empty Data Model Designer will be invoked. If you selected Map to an existing database, the following page is displayed instead.
The wizard displays a list of tables that can be mapped to persistent classes. Select a table (or tables) to be mapped to a persistent object (or persistent objects), and for each table select the columns that will be mapped to the persistent object’s properties. Persistent properties will not be generated for unchecked columns.
Note
If a table or column is displayed in gray and cannot be selected, the wizard is unable to import it. Move the mouse pointer over the disabled item to see the reason for this in a tooltip.
On the next page, you can customize the data model settings.
Class and property names in the generated code are the same as their mapped table and column names. You can add/remove custom prefixes and suffixes, and change the case style of names (e.g., capitalize names automatically). You can preview the resulting names immediately, and manually edit them later in the Data Model Designer. The correct mapping will be preserved as the PersistentAttribute with the PersistentAttribute.MapTo parameter is automatically applied in the underlying code.
The Nullable behavior option specifies whether or not nullable fields should be generated for columns depending on their (NOT) NULL flag. The following table shows the IsNulllable property value of generated fields for different column types.
| Database column type | AlwaysAllowNulls | ByUnderlyingType |
|---|---|---|
| Primary key column | false | false |
| Foreign key column | false | false |
| Data column with NULL flag | false | true |
| Data column with NOT NULL flag | false | false |
The Generate comments option enables the addition of comments to the data model. If an issue occurs during model generation, a descriptive comment is automatically added to the problematic model element. Created comments can be viewed later in the Data Model Designer.
The Create partial declarations for manual editing option adds the partial keyword to the class declaration, and splits each class implementation into two files (e.g., MyObject.cs and MyObject.Designer.cs). Files with the “Designer“ suffix are handled by the designer and should not be edited manually, as the designer will override your changes. Instead, your code should be placed into the file without the designer suffix. When the Create partial declarations for manual editing option is disabled, persistent class declarations are not split, and a single file is created for each class.
The generated data model will be displayed in the Data Model Designer after completing the wizard.
Note
For ASP.NET Core projects, the Data Model Wizard also generates extension methods to support Dependency Injection and JSON Serialization.